tomosia_amanaplus_crawl 0.1.8 → 0.1.9

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA256:
-  metadata.gz: 793268d3d7e8d3c1f3fd17ca9f1283839dda2a20dfb1a8769c65fd6d25bf0236
-  data.tar.gz: 115062f473eb2ca55c9ceea17265b73827b53feb548a48861b808b9baa64447f
+  metadata.gz: a3d62cab8ce5e9076b05594851cde7da03d627b8c386dfcbc57e64d561e7569a
+  data.tar.gz: 7ce113e8fde42d24405c662998c07cc6992a7d7e7ce94816590ef0a3f634ce0f
 SHA512:
-  metadata.gz: 83ee2ae24471817f907373dd2130fd54726786eb2fca7732e9fbe10ec841b4b5cbfc3fa4925753d725eb23362c330cec055a8b43eb8f0321053e7ce497f77930
-  data.tar.gz: 5b92ea21908fa8288a2fa26af61c8c198b525f0de6f6aba7766e89931f064edd0bd37059d2fc205b8090bf0aac2825865fa5fe9ce8226dd7003efb74f55ecf2b
+  metadata.gz: c4841b8f6948aa3a2afc7456380d26b374157ce5a453d9eb0bab6141ff5fb71b88cd3e6f51e0556cfb43421cef12c572a17a4a05f5d4d00404d19b8af452d079
+  data.tar.gz: 407ea07476e5fcbc4f356f540d975a4d15d36aab8ae02103e906575dc1153c257b2ada9f1d9309f39d9557e9109bfe464eae94680715143259af5b41ec6c4d74

data/Gemfile.lock

@@ -1,10 +1,11 @@
 PATH
   remote: .
   specs:
-    tomosia_amanaplus_crawl (0.1.5)
+    tomosia_amanaplus_crawl (0.1.8)
       httparty (= 0.18.1)
       nokogiri (= 1.10.10)
       spreadsheet (= 1.2.6)
+      thor
 
 GEM
   remote: https://rubygems.org/
@@ -37,6 +38,7 @@ GEM
     ruby-ole (1.2.12.2)
     spreadsheet (1.2.6)
       ruby-ole (>= 1.0)
+    thor (1.0.1)
 
 PLATFORMS
   ruby

data/README.md

@@ -23,9 +23,9 @@ Or install it yourself as:
 ## Usage
 
 ```ruby
-    require 'tomosia_amanaplus_crawl'
-    TomosiaAmanaplusCrawl::Crawler.new.run(keyword, path, max)
+    tomosia_amanaplus_crawl crawl "keyword" --destination "./lib" --max=123
 ```
+    Example: tomosia_amanaplus_crawl crawl "hoian" --destination "./" --max=123
     keyword: hoian, danang, ...
     path: './', '/desktop/', ...
     max: số lượng ảnh muốn lấy về. Nếu max lớn hơn tổng số ảnh các page thì vẫn lấy hết tất cả ảnh

data/Rakefile

@@ -4,3 +4,9 @@ require "rspec/core/rake_task"
 RSpec::Core::RakeTask.new(:spec)
 
 task :default => :spec
+
+namespace :gem do
+  task :build do
+    system "rake build && rake install"
+  end
+end

data/exe/tomosia_amanaplus_crawl

@@ -0,0 +1,5 @@
+#!/usr/bin/env ruby
+
+require 'tomosia_amanaplus_crawl/cli'
+
+TomosiaAmanaplusCrawl::Cli.start

data/lib/tomosia_amanaplus_crawl.rb

@@ -1,10 +1,6 @@
 require "tomosia_amanaplus_crawl/version"
 
 module TomosiaAmanaplusCrawl
-  def self.yeuNgucLep
-    puts "Yeu chi My nhieu lam"
-  end
-
   require 'nokogiri'
   require 'httparty'
   require 'open-uri'
@@ -12,7 +8,7 @@ module TomosiaAmanaplusCrawl
   require 'spreadsheet'
 
   class Crawler
-    URL = "https://plus.amanaimages.com/items/search/"
+    URL = "https://plus.amanaimages.com/items/search"
 
     def run(keyword, destination, max)
       unparsed_page = HTTParty.get("#{URL}/#{keyword}")
@@ -23,7 +19,7 @@ module TomosiaAmanaplusCrawl
     
       # lấy tổng số image
       total = parsed_page.css("h1.p-search-result__ttl").text.split(' ').first
-      total = total[11..(total.length - 1)].chop.chop.chop.sub(',', '').to_i
+      total = total[(6 + keyword.length)..(total.length - 1)].chop.chop.chop.gsub(',', '').to_i
       if max > total # nếu max lớn hơn total thì max = total => vẫn lấy hết
         max = total
       end
@@ -74,13 +70,23 @@ module TomosiaAmanaplusCrawl
       print "\nDownloading"
       images.each do |curr_image|
         threads << Thread.new(curr_image) {
-          open(curr_image[:url]) do |image|
-            File.open("#{path}/#{curr_image[:url].split('/').last}", "a+") do |file|
-              file.write(image.read) # lưu hình ảnh
-              curr_image[:size] = image.size # cập nhật lại size trong mảng images
-              print "."
+          timeout = 0
+          begin
+            open(curr_image[:url]) do |image|
+              File.open("#{path}/#{curr_image[:url].split('/').last}", "a+") do |file|
+                file.write(image.read) # lưu hình ảnh
+                curr_image[:size] = image.size # cập nhật lại size trong mảng images
+                print "."
+              end
+            end # end open
+          rescue => exception
+            if timeout < 3
+              timeout += 1
+              retry
+            else
+              next
             end
-          end # end open
+          end
         }
       end
       threads.each { |t| t.join }

data/lib/tomosia_amanaplus_crawl/cli.rb

@@ -0,0 +1,14 @@
+require 'thor'
+require_relative '../tomosia_amanaplus_crawl'
+
+module TomosiaAmanaplusCrawl
+  class Cli < Thor
+
+    desc "crawl KEYWORD", "enter KEYWORD to search"
+    option :destination
+    option :max
+    def crawl(keyword)
+      TomosiaAmanaplusCrawl::Crawler.new.run(keyword, options[:destination], options[:max].to_i)
+    end
+  end
+end

data/lib/tomosia_amanaplus_crawl/version.rb

@@ -1,3 +1,3 @@
 module TomosiaAmanaplusCrawl
-  VERSION = "0.1.8"
+  VERSION = "0.1.9"
 end

data/spec/spec_helper.rb

@@ -0,0 +1,14 @@
+require "bundler/setup"
+require "tomosia_amanaplus_crawl"
+
+RSpec.configure do |config|
+  # Enable flags like --only-failures and --next-failure
+  config.example_status_persistence_file_path = ".rspec_status"
+
+  # Disable RSpec exposing methods globally on `Module` and `main`
+  config.disable_monkey_patching!
+
+  config.expect_with :rspec do |c|
+    c.syntax = :expect
+  end
+end

data/spec/tomosia_amanaplus_crawl_spec.rb

@@ -0,0 +1,9 @@
+RSpec.describe TomosiaAmanaplusCrawl do
+  it "has a version number" do
+    expect(TomosiaAmanaplusCrawl::VERSION).not_to be nil
+  end
+
+  it "does something useful" do
+    expect(false).to eq(true)
+  end
+end

data/tomosia_amanaplus_crawl.gemspec

@@ -9,13 +9,13 @@ Gem::Specification.new do |spec|
   spec.homepage      = "https://github.com/tthuydang/tomosia_amanaplus_crawl"
   spec.required_ruby_version = Gem::Requirement.new(">= 2.3.0")
 
-  spec.files         = Dir.chdir(File.expand_path('..', __FILE__)) do
-    `git ls-files -z`.split("\x0").reject { |f| f.match(%r{^(test|spec|features)/}) }
-  end
-  spec.executables   = spec.files.grep(%r{^exe/}) { |f| File.basename(f) }
+  spec.files         = `git ls-files`.split("\n")
+  spec.bindir        = "exe"
+  spec.executables   = 'tomosia_amanaplus_crawl'
   spec.require_paths = ["lib"]
 
   spec.add_runtime_dependency('httparty', '0.18.1')
   spec.add_runtime_dependency('nokogiri', '1.10.10')
   spec.add_runtime_dependency('spreadsheet', '1.2.6')
+  spec.add_runtime_dependency('thor')
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: tomosia_amanaplus_crawl
 version: !ruby/object:Gem::Version
-  version: 0.1.8
+  version: 0.1.9
 platform: ruby
 authors:
 - Nhat Huy
 autorequire: 
-bindir: bin
+bindir: exe
 cert_chain: []
-date: 2020-08-07 00:00:00.000000000 Z
+date: 2020-08-08 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: httparty
@@ -52,9 +52,24 @@ dependencies:
     - - '='
       - !ruby/object:Gem::Version
         version: 1.2.6
+- !ruby/object:Gem::Dependency
+  name: thor
+  requirement: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :runtime
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    requirements:
+    - - ">="
+      - !ruby/object:Gem::Version
+        version: '0'
 description: 
 email: 
-executables: []
+executables:
+- tomosia_amanaplus_crawl
 extensions: []
 extra_rdoc_files: []
 files:
@@ -69,8 +84,12 @@ files:
 - Rakefile
 - bin/console
 - bin/setup
+- exe/tomosia_amanaplus_crawl
 - lib/tomosia_amanaplus_crawl.rb
+- lib/tomosia_amanaplus_crawl/cli.rb
 - lib/tomosia_amanaplus_crawl/version.rb
+- spec/spec_helper.rb
+- spec/tomosia_amanaplus_crawl_spec.rb
 - tomosia_amanaplus_crawl.gemspec
 homepage: https://github.com/tthuydang/tomosia_amanaplus_crawl
 licenses: []